326,7,21�3$3(5�21�0(5&85<� &+$37(5��

1-23

&+$37(5��� (;32685(�72�0(5&85<�,1�(8523(

���� ,1752'8&7,21

���� 6285&(6�$1'�3$77(516�2)�(;32685(

5.2.1 AIR

5.2.2 WATER

5.2.3 SOIL

5.2.4 DIET

���� %,2021,725,1*�2)�0(5&85<�(;32685(�,1�+80$16

5.3.1 ELEMENTAL AND INORGANIC MERCURY

5.3.2 METHYLMERCURY

5.4 $1$/<7,&$/�$1'�0($685(0(17�0(7+2'6�)25�%,2/2*,&$/�021,725,1*

5.4.1 DETERMINATION OF TOTAL MERCURY

5.4.2 INSTRUMENTAL METHODS

5.4.3 DETERMINATION OF ORGANOMERCURY COMPOUNDS

5.4.4 QUALITY CONTROL AND QUALITY ASSURANCE

���� 0($685('�$1'�02'(//('�/(9(/6�2)�0(5&85<�,1�%,20$5.(56

5.5.1 MODELLING HUMAN EXPOSURE

5.5.2 MEASURED BIOMARKERS OF EXPOSURE

���� &21&/86,216�$1'�5(&200(1'$7,216

���� 5()(5(1&(6

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

2-23

��� ,1752'8&7,21

Although the question on what is the acceptable level of exposure to MeHg and otherchemical forms of mercury is not new to the scientific community, there are still a lot ofcontroversies in this subject, which become even more evident in recent studies. Although itwas believed for a long time that hair and blood are appropriate biomarkers or measurementsof exposure, there are controversies in recent studies whether a measured biomarker valuetruly corresponds to Hg exposure levels. This chapter provides a short overview on thesubject and provides suggestions for further studies to improve uncertainties in modelpredictions of individual and population exposures to mercury. In the limited scope of theposition paper only a very brief summary is provided based on recent reviews (NRC 2000, USEPA 1997) as well as environmental health criteria provided by the WHO (IPCS, 1990,1991).

��� 6285&(6�$1'�3$77(516�2)�(;32685(

There is a considerable variation of mercury levels in those media that are the sources ofhuman exposure, and, consequently, in their contribution to the toxicity risk. Main mercuryabsorption routes in humans are through respiratory and dietary routes. Non-occupationalgroups are primarily exposed through the diet and dental amalgam.

������� $PELHQW��$LU

In Europe the highest background concentration of TGM is measured in central Europe (e.g.Germany and Poland), where the concentrations may reach up to 2.5 ng/m3 (EMEP; 1999).The concentrations in rural areas are normally very low, close to the mean global values(EMEP, 1999). The values in urban areas are usually higher and vary between 5 to 15 ng/m3

(IPCS, 1991) and in some contaminated places even higher ('L]GDUHYLþ���������7KH�UHIHUHQFHconcentrations (RfC) recommended by the US EPA amounts to 0.3 µg/m3 (IRIS, 1995),which means that in general mercury concentrations in air do not represent a considerableintake of Hg for humans. The WHO has estimated the daily intake of each form of Hg on theassumption that 75% of Hg is in elemental Hg from, 5% as inorganic Hg and 20% of MeHg.By assuming a daily ventilation of 20 m3, and the amount absorbed across the pulmonarymembranes (80% of elemental Hg, 50’% of inorganic Hg, and 80% of MeHg) daily intakeswere calculated and given in Table 5.1.

������ 'HQWDO�$PDOJDP

Dental mercury fillings are reported to release Hg vapour into the oral cavity (Clarkson et al.,1988 a, 1988b, Skerfving, 1991). The resulting concentrations in intra-oral air cansubstantially exceed those found in the ambient atmosphere, especially after a period ofchewing. It is estimated that average daily amounts of Hg entering the pulmonary systemranges from 3 – 17 µg of Hg (IPCS, 1991, Berglund 1990, Lorscheider 1995, Sandborgh

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

3-23

1998, Barregard et al. 1995, Sallsten et al., 1996), but there is variability among populations,and intakes in the order of 100 ug/day may occur.

7DEOH���� - Estimated average daily intakes and retention in (µg/day) ofdifferent mercury forms in the general population not occupationally

exposed (IPCS, 1991).

([SRVXUH (OHPHQWDO�+JYDSRXU

,QRUJDQLF�+JFRPSRXQGV

0HWK\OPHUFXU\

Air 0.03(0.024)* 0.002(0.001) 0.008(0.0069

Dental amalgams 3.8-21 (3-17) 0 0

Food

- fish

- non-fish

0

0

0.60 (0.042)

3.6 (0.25)

2.4 (2.3)**

0

Drinking water 0 0.050 (0.0035) 0

Total 3.9-21 (3-1-17) 4.3 (0.3) 2.41 (2.31)

The data in parenthesis represent retained Hg in the body of an adult.* If the concentration is assumed to be 15 ng/m3 in an urban area the figure would be

0.3(0.24) µg/day** 100 g of fish per week with the Hg concentration of 0.2 mg/kg.

����� :DWHU

Concentration of mercury in drinking water is normally very low (below 1ng/l). Values of upto 25 ng/l are reported (IPCS, 1991). Based on the assumption that an adult consumes about 2l of water per day, the daily intakes of Hg from drinking water is insignificant. The WHOguidelines and many national legislation set the values of 1000 ng/l (WHO-FAO, 1993).

������ 6RLO

The average mercury concentration in surface soil is reported to be from 20 to 625 µg/kg(Reiman and de Caritat, 1998). Higher concentrations are reported in soils from urbanlocations and close to sources of Hg pollution (smelting, mining, coal burning facilities, chlor-alkali industry, etc). In Europe very little knowledge is available for vitalization of Hg fromsoil and consequently on the direct exposure of humans to Hg in soils.

������ 'LHW

Daily intakes and retention of mercury from food is difficult to estimate accurately. In mostfood stuff Hg concentration is below 20 µg/kg. Mercury is known to bioconcentrate in aquatic

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

4-23

organisms and it is biomagnified in aquatic food webs. For example, the concentration of Hgin small fish at low food web level (such as anchovies) is below 0.085 mg/kg, while in swardfish, shark and tuna values above 1.2. mg/kg are frequently reported (IPCS, 1991). InScandinavian predatory fresh-water fish (perch and pike) average levels are about 0.5 mg/kg.

The use of fish meal as the feed for poultry and other animals used for human consumptionmay results in increased levels of Hg. In Germany, the poultry contains 0.03 - 0.04 mg/kg.Cattle are able to demethylate Hg in the rumen, and therefore, beef meet and milk containvery low concentrations of Hg.

7DEOH����� Current data on the daily dietary intake of mercury

&RXQWU\ ,QWDNH�(µg day-1) 5HIHUHQFHV

Belgium All food: ���of which ����is fromfishAll foodstuff: ���

Fouassin and Fondu, 1978Buchet et al. 1983

Poland ���� ( age group 1-6 years)���� (age group 6-18 years)�����in adultsFrom fish: 7% of total dietaryintake

Szprengier-Juszkiewicz,1988

Nabrzyski and Gajewska,1984

Germany ��� from fish��� from food (except fish andvegetables)

LAI, 1996

Croatia From fish: ������total Hg)������ MeHg form)

Buzina et al., 1995

Spain �����60-90 % from seafood)in Velencia only 27% is from theseafood���of which about �� is from fish(Bask country)

Moreiras et al. 1996

Urieta et al., 1996

Sweden ����(market-basket) Becker and Kumpulainen,1991

United Kinghdom � MAFF, 1994Finland � Kumpulainen and Tahvonen,

1989The Netherlands ��� Van Dokkum et al., 1989Czech Rep. ��� Ruprich, 1995

Brazil ���� �� � ���� (Amazon, Medeirariver)

Boishio and Henshel, 1996

Japan ������±�����������as MeHg�

Tsuda et al., 1996Ikerashi et al., 1996Nakagawa et al., 1997

The main problem to accurately estimate daily intakes of various Hg forms from diet is thatnational survey programmes mainly report total Hg concentrations and the percentage of Hgas MeHg is not known. Total mercury daily intakes reported in various countries are given inTable 5.2. In some national survey the percentage of Hg originating from fish is provided. It

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

5-23

is assumed that in this foodstuff the percentage of as MeHg is from 60 to 90 %. Therefore fishand fish products represent the major source of methylmercury.

The US EPA reference dose (RfD) for MeHg is 0.1 µg/kg body weight/day (IRIS, 1995a).This would be 42 µg/week for a 60 kg adult. The equivalent amount of fish consumed wouldbe 420 g of fish per week with 0.1 mg Hg/kg or 105 g of fish per week with 0.4 mg Hg /kg.WHO-FAO provisional tolerable weekly intake (PTWI) amounts to 5 µg/kg body weight (0.7µg/kg body weight/day), but a revision is currently in progress. The PTWI set by the WelfareMinistry of Japan is 0.17 mg of methylmercury (0.4 µg/kg body weight/day) (Nakagawa etal., 1997). In Europe the recommended limits for mercury vary. In Scandinavian countries Hgin fish should not exceed 0.5 mgHg/kg. The Swedish Food and Health Administration is atpresent evaluating the relevance of a new limit at 0.3 mgHg/kg, while Japan has alreadyadopted a 0.3 mg/kg guideline (Dickman and Leung, 1998). In Sweden pregnant women areadvised not to eat perch an pike at all during pregnancy.

It may be concluded that in those areas in Europe where fish consumption represent aconsiderable part of diet, the value the US EPA RfD could be considerably exceeded.

��� %,2021,725,1*�2)�0(5&85<�(;32685(�,1�+80$16

This chapter deals with biological samples that are used to assess mercury exposure inhumans. The selection of biological media depends on mercury compounds, exposure pattern(e.g. chronic, acute) and time of sampling after exposure.

����� (OHPHQWDO�DQG�,QRUJDQLF�0HUFXU\

In case of exposure to elemental mercury blood and urinary mercury are commonly used toassess occupational exposure. Elemental Hg° in exhaled air and urine has also been used toassess the level of recent exposure to elemental Hg. Kinetics of elemental Hg and biologicalsamples for biomonitoring are presented in Figure 5.1. One should note that inhaled vapour ofHg is oxidised to Hg (II) and both species are present (elemental and divalent Hg). Theelemental Hg is highly mobile, readily crossed the placenta, cell membranes, and the blood-brain barrier. The Hg (II) ions are much less mobile, crossing the above barriers at a muchslower rate.

The usefulness of blood as a bioindicators of exposure to elemental Hg depends on timeelapsed since exposure and the level of exposure. The whole blood analysis may be used toassess the exposure. Mercury in blood increase rapidly with the exposure, and decreases withan initial half-life of approximately two to four days, and a slower phase of a couple of weeks(Barregard et al. 1992, Sallsten et al. 1993). This means, that the usefulness of the blood is oflimited value, in particular, if it is taken several weeks after exposure. It is, however, valuablefor assessment of peak exposures (Barregård 1993).

At low levels of elemental Hg exposure, individual differences in total Hg in blood could beexplained by current exposure to Hg due to the number of amalgams fillings, fishconsumption and other possible exposure routes (e.g. living environment in Hg mining area).

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

6-23

At low-level elemental Hg exposure, blood Hg poorly represent information on current andpast Hg° exposure. A separation of whole blood into its plasma and erythrocytes fractionspermits better discrimination between exposure to Hg0 and methyl mercury. Suchdiscrimination will, of course, be more correct using speciation of mercury in blood.

,QJHVWLRQ6NLQ�$EVRUSWLRQ,QKDODWLRQ

LungLung Skin( )HSLWK��FHOOV

Gl-tract

Blood

Vapour Vapour(Hg )++

Liver

CNS

Fetus

Kidney

Urine Feces

Hg++

Critical organs

Media for biological monitoring

Hg++

Exfoliatedcells and

sweat

Placenta

ExhaledAir

Vapour

Accumulation

Possible media for biological monitoring

)LJXUH�����-�Metabolism and media for biological monitoring(Elinder et al. 1988)

Urinary Hg indicates mercury levels present in kidneys (Clarkson et al. 1988a,b). In the caseof occupational exposure, urinary Hg has been used to estimate exposure. In urine, mercurylevels decrease at a much slower rate as compared to blood. The half live of exposure variesbetween 40 and 90 days (Barregård et al. 1992, Roels et al. 1991, Sallsten et al. 1994). Afterexposure to higher levels of mercury, urinary Hg levels stay elevated over a longer period oftime, which means that urine is a much better indicator for longer periods of exposure thanblood samples. Internal dose of mercury is normally assessed using determination of mercuryin first morning urine, corrected for varying dilution using creatinine determination. Urinarymethylmercury levels are normal very low, which means that very little dietary Hg is excretedby urine. It is important to note that good correlation between urinary Hg values andconcentration of Hg in air has only been demonstrated after stable exposure and correction ofthe urinary Hg values with the urinary excretion rate and normalization to the time elapsedafter exposure (Roels et al. 1987). However, intra-individual differences still remain high. It isinteresting to note that Hg levels in former Hg miners and active miners currently notexposed, Hg levels in urine are comparable to those of occupationally non-exposedpopulation (.REDO�HW�DO���������.REDO�DQG�'L]GDUHYLþ��������

The concentration of Hg in urine is a good indicator of a long-term integrated exposure, whiletotal Hg in blood is a good indicator of recent exposure (Gompertz, 1982, Barregård 1993).After inhalation mercury is rapidly oxidized, bound to organic molecules, and is then excretedin the bound form through urine. At a higher rate of absorption, a small part of the Hgcirculates in the blood and kidney in elemental Hg form, and is partially excreted as dissolvedelemental Hg in urine (Yamamura, 1991).

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

7-23

An attempt has been made to use the exhaled air as a possible bioindicators of exposure toelemental Hg (Sallsten et al. 1998, 2000). A portion of absorbed Hg is excreted via the lungs.Excretion half-life is approximately 18 hours (Hursh et al. 1976). At a lower levels ofexposure, the usefulness of this biomarker is limited due to numerous confounding factors,such as Hg released form he amalgam filling, drinking of alcohol, etc..

The kinetics of uptake and release in blood and urine are fairly rapid, while Hg levels in targetorgans may change far more slowly (Clarkson, 1988). Mercury levels in blood and urine maynot reflect the accumulation of Hg in the brain, while urinary Hg mainly reflects kidney Hg(Carkson 1988 a,b, Borjesson et al. 1995). The mobilisation tests with 2,3-dimercaptopropane-1-sulfonate (DMPS) or meso-2,3-dimercaptosuccinic acid (DMSA)increases the amount of Hg excreted, but does not yield any additional information to pre-chelation excretion (Sallsten et al. 1994).

As the elemental Hg vapour is readily oxidized to inorganic Hg after absorption, similarbiomarkers (blood and urine) could be used to assess exposure to inorganic Hg.

������ 0HWK\OPHUFXU\

In case of exposure to methylmercury two bioindicators are normally used: blood and hairsamples. In the blood MeHg accumulates mainly in the red blood cells, therefore Hg in redblood cells is frequently used as an index to MeHg exposure. Exposure to elemental Hg willled to an increased values of Hg in plasma, therefore through the analysis of total Hg in thesesamples it is possible to differentiate between exposure to elemental Hg and MeHg. However,it is recommended to measure total Hg and MeHg in order to take into account possible co-exposure to elemental/inorganic Hg. This is not difficult as a number of methods have beendeveloped in recent years to facilitate accurate analysis of total and MeHg in biologicalmaterials. The kinetics of MeHg in humans and corresponded tissues suitable forbiomonitoring are presented in Figure 5.2.

Scalp hair has widely been used as a good indicator of exposure MeHg in the diet. MeHg isincorporated into hair follicle in proportion to its content in blood.

The hair-to-blood ratio in humans has been estimated as approximately 250:1 expressed asµg/g hair to mgHg/L blood. Once MeHg is incorporated into hair, MeHg is stable, andtherefore provides a longitudinal history of blood MeHg. Hair grows with an approximaterate of about 1 cm/month, recapitulation over a long period is possible, dependent on thelength of the hair. In population constantly exposed to MeHg, the percentage of Hg as MeHgis close to 100%. However, due to possible co-exposure to elemental Hg or possible externalcontamination of hair samples with inorganic Hg, it is recommended that both total Hg andMeHg are measured, in cases when high levels of Hg are found.

Concentrations of MeHg in body and/or pubic hair were as well proven to be good indicatorsof MeHg burden (Horvat et al., 1988).

It is well understood that pregnant women and their foetus are the most critical populationgroup as regards the exposure to MeHg through food consumption. Some recent studies

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

8-23

carried out in Faroe islands (Grandjean et al., 1997, Sorensen et al. 1999, Budtz-Jorgensen etal. 2000, Steuerwald et al. 2000) indicated the usefulness of umbilical cord blood as goodexposure indicator during pregnancy. There is a good correlation between MeHg in maternaland umbilical cord blood, with higher values found in cord blood, indicating efficienttransport of MeHg through the placental barrier to the foetus. At more elevated levels ofMeHg in maternal blood (about 7 ng/ml), the umbilical: maternal blood ratio is about 3. Theusefulness of placenta as an indicator of exposure is still not clear, although most studiesshow very good correlation between MeHg in blood, hair and placenta. The percentage of Hgand MeHg in placenta is much lower than in blood samples, most probably reflecting removalof inorganic Hg before it can be transferred to the foetus (Horvat et el. 1988).

,QJHVWLRQ6NLQ�$EVRUSWLRQ,QKDODWLRQ

LungLung Skin Gl-tract

Blood

High High

Demethilation?

Liver

CNS

Placenta

Fetus(CNS)

Kidney

Urine Feces

Hg++MainlyHg++ Hair

MeHg

-95%

Critical organs

Media for biological monitoring

)LJXUH�����- Metabolism and media for biological monitoring of methylmercury(Elinder et al. 1988)

The poisoning nature of mercury is well acknowledged (IPCS 1991; Ishihara andUrushiyama, 1994). But less known are the effects of mercury on humans as a consequence oflong term exposure to low concentrations. In many cases the use of biomarkers, such as Hgconcentrations blood and urine, are not sufficient to assess the internal doses and potentialeffects on the central nervous system, kidney, the immune system, and other possible effects.Therefore, better scientific understanding of risks to human health, especially to those citizensliving close to potentially dangerous sites, is needed. Therefore other biomarkers thanmercury measurements alone should be used. An example is N-acetyl-glucose-aminidase(NAG) and other low molecular weight proteins in urine, that seem to reflect effects at lowlevel exposure to Hg0. There is a need for continuous research, and for example, markers ofoxidative damage could be tested.

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

9-23

���� $1$/<7,&$/�0(7+2'6�)25�%,2/2*,&$/�021,725,1*

Methods for the determination of total and major species of mercury are classified accordingto the isolation techniques and detection systems. They are selected depending on the natureof the sample and in particular the concentration levels of mercury. The key elements forobtaining accurate measurements are connected with contamination-free sampling, samplestorage and handling. The analytical methods for determination of total mercury and itsspecies in various biological and environmental samples and the needs for future developmentare reviewed (Horvat, 1996; Horvat and Schroeder, 1995; Schroeder, 1995).

Relatively little is known about the effects of storage on the stability of total Hg andmonomethylmercury compounds (MeHg) in biological samples. Fresh samples are usuallystored (deep frozen or lyophilised) in darkness (Horvat and Byne, 1992). Sometimes samplesare sterilized (ϒ-irradiation or autoclaving) in order to prevent further transformation ofmercury species by bacteriological activity. Sample preparation must be carried out underclean mercury-free conditions. Significant external contamination of samples withmethylmercury is unlikely to occur, however extreme cautions are necessary to avoidcontamination by inorganic mercury.

������ 'HWHUPLQDWLRQ�RI�7RWDO�0HUFXU\

Most of the methods for the determination of total mercury in biological samples requirepreliminary digestion of the sample. They are classified as wet oxidizing digestion, forexample see Liang and Bloom (1993) and dry combustion/pyrolysis decomposition methods,for examples see (Byrne and Kosta, 1974; LECO, 1999).

����� ,QVWUXPHQWDO�0HWKRGV

Frequently applied detection techniques are cold vapour atomic absorption spectrometry (CVAAS), more sensitive atomic fluorescence spectrometry (CV AFS), and various types ofemission spectrometry. Their relative detection limits are presented in Table 5.3.

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

10-23

7DEOH���� - Most frequently used methods for quantification ofmercury and their relative detection limits (Horvat, 1996)

0HWKRG 5HSRUWHG�'HWHFWLRQ/LPLWV

Colorimetric methods 0.01 - 0.1 µg/gAAS graphite furnace (GF AAS) cold vapour (CV AAS)

1 ng/g0.01 - 1 ng/g

AFS cold vapour (CV AFS) 0.001 - 0.01 ng/gNAA instrumental (INAA) radiochemical (RNAA)

1-10 ng/g0.01 - 1 ng/g

GC Electron Capture Detector Atomic Emission Detector Mass Spectrometer CV AAS/AFS

0.01 - 0.05 ng/g ~ 0.05 ng/g0.1 ng/g0.01- 0.05 ng/g

HPLC UV CV AAS CV AFS Electrochemical detectors

1 ng/ml0.5 ng/ml0.08 ng/ml0.1-1 ng/ml

ICP-MS 0.01 ng/mlICP-AES 2 ng/mlPhoto-acoustic spectroscopy 0.05 ng

X ray fluorescence 5 ng/g - 1 µg/gElectrochemical methods 0.1 - 1 ng/gGold-film analyzer 0.05 µg/g

����� �'HWHUPLQDWLRQ�RI�2UJDQRPHUFXU\�&RPSRXQGV

Hundreds of papers on the analyses of organomercury compounds in environmental sampleshave appeared during the past twenty years. They are systematically reviewed in severalarticles and monographs (Rodriquez-Vazquez, 1978, Craig, 1982, Horvat and Schroeder,1995, Horvat, 1996). In general, methods are classified according to the isolation techniqueand the detection system. Most common steps for determination of MeHg are reviewed andschematically presented in Figure 5.3.

There are also a few methods that are based on differential reduction, which is frequently usedin clinical laboratories. In the method developed by Magos (Magos, 1971) the inorganicmercury in an alkaline digested sample is selectively reduced by stannous chloride whileorganomercury compounds are reduced to elemental mercury by a stannous chloride-cadmium chloride combination. CV AAS can measure elemental mercury released.

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

11-23

6DPSOH�FROOHFWLRQ↓

6DPSOH�SUHWUHDWPHQW↓

/LEHUDWLRQ�RI�0H+J�IURP�LWV�PDWUL[(acid leaching, alkaline dissolution, volatilisation, distillation,

superfluid extraction, microwave assistance)↓

([WUDFWLRQ�FOHDQ�XS�SUHFRQFHQWUDWLRQ(Solvent extraction; derivatization such as ethylation, butylation, hydration and

iodination;cryogenic trapping; preconcentration on solid phases)

↓6HSDUDWLRQ�RI�PHUFXU\�VSHFLHV�RI�LQWHUHVW(gas chromatography; HPLC, ion-exchange)

↓4XDQWLILFDWLRQ

(CV AAS, CV AFS, GC-ECD, AED, MS, ICP-MS)

CVAAS – cold vapour atomic absorption spectrometryCVAFS - cold vapour atomic fluorescence spectrometryGC-ECD - gas chromatography coupled with the electron capture detectorAED - atomic emission spectrometryICP-MS – inductively coupled mass spectrometryHPLC – high performance liquid chromatography

)LJXUH���� - Steps for determination of MeHg

����� 4XDOLW\�&RQWURO�DQG�4XDOLW\�$VVXUDQFH

There are various means to achieve good quality data in analytical laboratories.Representative samples need to be properly collected and stored, prior to analysis, and alllaboratory work conducted under a good QA programme. This can efficiently be achievedthrough skilled, well-trained, experienced and motivated staff. Protocols for sampling andsample storage should be well developed in order to prevent contamination and /or losses ofHg. In addition, interconversion of various Hg species during sample handling should also beprevented. Samples should be processed under very clean laboratory conditions and theappropriate quality labware and reagents (Hg-free) should be employed. The analysts shouldonly use an analytical procedure on a routine basis after it has been validated for the range ofconcentrations and matrices to be dealt with the measurement programme using the relevantreference materials (RMs) and certified reference materials (CRMs). Other importantmeasures include the quality assurance manual, training of personnel, a good managerialstructure of the laboratory, the use of validated methods, the application of statistical controlprinciples (e.g. control charts), and the external quality control measures (e.g. interlaboratorytests).

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

12-23

There are a number of CRMs available for total mercury in biological, sediment, and watersamples (IAEA, 1995, 1996). In addition, the COMAR database(http:www.bam.de/a_i/comar/scr/titel.htm) is one of the most extensive data base on theavailability of certified reference materials on a global scale and it is continuously updatedHowever, only human hair CRMs are available for determination of both, total Hg and MeHgin samples of human exposure (Horvat, 1999).

Currently available CRMs are not sufficient to cover the needs in terms of various origin(urine, blood, hair) and concentration ranges. Human hair CRMs will soon become obsoleteand to our knowledge no new CRMs are in preparation. In the absence of proper CRMs analternative way to control the accuracy of analytical data is regular participation inintercomparison exercises.

��� �0($685('�$1'�02'(/('�/(9(/6�2)�0(5&85<�,1�%,20$5.(56

����� 0RGHOOLQJ�+XPDQ�([SRVXUH

Humans are exposed to mercury of natural and anthropogenic sources. Models weredeveloped to calculate exposure of individual and population exposures. In Europe suchlarge-scale studies are practically non-existence, however in the USA various scenarios weremodelled (US EPA, 1996), although large uncertainties exists from one to anothergeographical locations.

Models consist of various sub-models mainly dealing with simulation of environmental fateof mercury (Table 5.4.).

7DEOH�����- Models used in US to predict mercury air concentrations,deposition fluxes and environmental concentrations (US EPA, 1997).

Atmospheric model Predicts average annual atmospheric mercury concentrationand wet and dry deposition flux for selected grid, due to allanthropogenic sources of mercury and a natural backgroundatmospheric Hg concentrations. It also accounts for the long-range transport of mercury emitted from anthropogenicsources.

Local air transportmodel (GAS -ISC3)

Calculates average concentrations and deposition fluxeswithin a selected grid from emission source (categorised infour major groups scenarios)

Modellingenvironmentalconcentrations(IEM-2M)

Predicts environmental concentrations of three chemicalforms of Hg (elemental, divalent and monomethylmercurycompounds) in soil, water and biota. It is based on local-scale estimates of air concentrations and deposition rates towatersheds and water bodies.

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

13-23

Model for human exposure through inhalation and ingestion of other contaminated food werealso evaluated. Atmospheric concentrations are predicted by atmospheric models and soilconcentrations from the IEM-2M model. Concentrations in plants are calculated from soil-toplant and air-to plant transfer factors. Concentrations in animal tissue are then calculated frompredicted concentration in plants, animal consumption rates, and specific biotransfer factors.Mercury concentrations in fish are calculated from simplified assumptions that aquatic foodchains ca be adequately represented using four trophic levels: phytoplankton (algalproducers), zooplankton (primary herbivorous consumers), small forage fish (secondaryconsumers) and larger, piscivorous fish (tertiary consumers), which are consumed by humans.It is well documented that food chain structures vary considerably among different aquaticsystems with large variability in bioaccumulation factors. Another important simplification inthese models is that the concentration of MeHg in fish is directly proportional to dissolvedMeHg concentrations in the water. It is also well documented that this relationship may varyconsiderably from one to another water body.

3UREDELOLW\�'LVWULEXWLRQ 3UREDELOLW\�$QDO\VLV

)LJXUH���� - Schematic description of the methodology for the probabilisticassessment of regional Hg exposure (Seigneur et al., 1997).

Regional Hgdeposition

fluxes

Hg loading tolake

Hgconcentrationsin lake water

MeHgconcentration

in fish

MeHg dose tohuman

population

Atmospheric ModelDeposition Fluxes

Simulations for several lakes

Fish consumption rate

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

14-23

The bioaccumulation factor (BAF) for the piscivorous fish is calculated to be 1.6.106,calculated on the assumption that the percentage of Hg as MeHg is about 7.8% of totaldissolved Hg in water, and 65% of these is freely dissolved.

Exposure scenarios are defined by the source of contamination, the potential receptorpopulation, the potential pathways of exposure and the variables that affect the exposurepathways. The fate of deposited mercury should be examined for each situation separately.This is of particular importance in case of expected elevated mercury concentration close toemission sources. The exposure scenarios must then include the total amount of food derivedfrom the affected area and the extend of mercury contamination of these food sources. Eachsituation should be examined separately in order to identify the most important factors thatshould be taken into account in the modelling, and by this to reduce uncertainties of simulatedresults.

A probabilistic assessment of regional mercury exposure was developed by Seigneur et al.(1997). The methodology is based on a multimedia model of the atmospheric fate andtransport of mercury over a continental scale and the aquatic simulation of the transformationand bioaccumulation of mercury in lakes in the region of interest. The model, however,addresses only the variabilities if key input variables and does not include a treatment of inputdata uncertainties or model formulation uncertainties. The methodology is schematicallypresented in Figure 5.4 and was practically applied in the Great Lakes region of the USA asan example.

������ 0HDVXUHG�%LRPDUNHUV�RI�([SRVXUH

Reference values for total mercury concentrations in biological media for the generalpopulation were discussed by the WHO (IPCS, 1990), in this context referred to as ‘normallevels’. The mean concentration for whole blood was considered to be about 8 µg/l. This levelseems to be too high as an average for the European populations of today. As the exposure tomethyl mercury predominantly is related to fish consumption the whole blood referencevalues will differ according to dietary habits, mainly fish consumption. In subjects with dentalamalgams and low exposure to methyl mercury, the amalgams will, however, also contributeto whole blood mercury levels (Clarkson et al. 1988b, Barregård 1993). In rare cases theblood mercury levels due to dental amalgam may be as high as 20 µg/l (Barregård et al.1995).

Scandinavian studies in the 1980-1990 showed average mercury levels in whole blood of 3-5µg/l in occupationally unexposed adults, used as referents in occupational epidemiologystudies (Akesson 1991, Langworth et al. 1991, Svensson et al. 1992, Barregård et al. 1994,Ellingsen 1993). Nearly all of them had amalgam fillings, and the average consumption of(salt water) fish was 1/week. In subjects born in the 1970s and later, the levels are lower,mean 1-2 µg/l owing to less dental amalgam and lower fish consumption (Sandborgh-Englundet al. 1996, 1998, Vahter et al. 2000). German adults have mean blood mercury levels of onlyabout 0.5 µg/l (Schweinsburg and Kroiher 1994, Drexler et al. 1998, Seifert et al. 2000)

Mean urine mercury levels in occupationally unexposed adults in Scandinavia were 2-3 µg/gcreatinine (2-4 µg/l in 24h urine) in the 1980s (Barregård 1993). In younger subjects, the

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

15-23

levels are lower, mean 1-2 µg/g creatinine, owing to less dental amalgam (Herrstrom et al.1995, Vahter et al. 2000). In Germany the levels are lower, about 0.5 µg/l (Zander et al. 1990,Drexler et at 1998, Ganss et al. 2000, Seifert et al. 2000).

For scalp hair the mean concentration was considered to be about 2 µg/g by the WHO. Thislevel too seems to be too high as an average for the European populations of today. In Swedenand Germany average hair mercury levels in adults are about 0.5 µg/g (Oscarsson 1994, 1996,Bratel 1997, Schweinsburg and Kroiher 1994).

In order to derive the reference doses for humans, it is necessary to determine the ingesteddose that resulted in the Hg concentration in the biomarker. In order to do such calculations itis necessary to use toxicokinetics models. There are several toxicokinetic parameters thatdetermine the tissue (or biomarker) MeHg concentration after ingestion of a given dose ofMeHg. For example in case of MeHg ingestion through food, those parameters include theuptake of MeHg from the gastrointestinal tract, the distribution of MeHg to the various bodytissues, and the elimination of MeHg or Hg from those tissues (Figures 2 and 3). Two modelsare used for this purpose. (1) physiologically based pharmacokinetic model - PBPK (Clewellet al. 1999) and (2) a single compartmental pharmacokinetic model (IPCS 1990; US EPA1997). The PBPK model attempts to characterise the distribution and redistribution of MeHgamong several body compartments, including maternal hair and foetal cord blood. This modelis conceptual more accurate and flexible compared to one-compartmental model; it is verycomplex and more difficult to evaluate.

The accumulation and excretion of MeHg in humans, measured in terms of hair and bloodlevels, can well be represented by single-compartment model. This is a useful working modelfor comparing blood and hair levels to daily intakes of methylmercury. This model collapsesthe maternal-body compartments into a single maternal blood compartment. Toxicokineticcompartments that are not directly considered in the model include MeHg in cord blood andother foetal tissues, net absorption of MeHg and MeHg in maternal hair.

In adult humans the whole body concentration would assume steady state in about one year.The comparison of predicted relationship with those observed in field studies on populationbelieved to have attained the steady state from long-term dietary exposure to MeHg in fishshows that field studies always show lower values. One reason may be that the populationsunder investigation were not in true steady state, since intake is frequently seasonal in fish-eating population. This model is refered to the average adult human of a body weight of 70kg. It is assumed that children and teenagers are more vulnerable to high intakes of MeHg dueto higher energy intake.

At a steady state the accumulated amount (A) is given by:

Where:

A = the accumulated amounta = the amount taken up by the body (or organ) dailyb = the elimination constant (related to the biological half-time)

ba

=A

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

16-23

The steady state concentration in blood may be related to the following equation (IPCS,1990):

Where:

C = Concentration in blood (µg/litre)0.95 = Fraction of the daily intake absorbed0.05 = part of absorbed amount that goes to the blood compartmentd = average daily dietary intake (µg of Hg)5 litres )= blood volume0.01 days-1 = elimination constant

Both models suffer from interindividual variability in physiology and kinetics, for which nocorrect value exist. Each of the model parameters is a random variable whose possible valuesin a population can be described by probability distribution. Failure to considerinterindividual toxicokinetic variabilities can result in an RfD that is not protective of asubstantial portion of the population. Interindividual toxicokinetic variability can beaddressed in the derivation of RfD by application of uncertainty factors to a central-tendencyestimate of the ingested dose. It is still not certain which values are most appropriate for themodel parameters used to derive the central tendency estimates. The choices for each caseshould be carefully considered with reference to discussions presented in the known andpublished analysis of toxicokinetic variability (NRC, 2000).

Concentration of MeHg in hair is proportional to blood concentration at the time of formationof the hair strand. In general, the concentration in hair is 250 times the simultaneousconcentration in blood. Once MeHg is incorporated in hair mercury concentration remainsunchanged. The mean mercury concentrations in hair and consumption of fish are presentedin Table 5.4.

An interesting example in the Mediterranean was conducted by the WHO and FAO whereover 4000 people were screened through a dietary survey, and a total of 1098 hair sampleswere analysed (659 from Greece, 241 from Italy, and 198 from Croatia) (WHO, 1988). Theresults confirmed a positive correlation between seafood consumption and the levels of MeHgin hair.

Based on the WHO criteria for identifying population possible at risk (25 µg/kg of Hg in hairof adults and 6 µg/kg in hair of pregnant women) none was identified in Croatia. In Greekpopulation only the residents of remote fishing communities, exceeded the criteria. Most ofthe Italian individuals have also been considered free from risk, except population offisherman spending considerable time at sea, and babies that had high intrauterine exposurefrom the high seafood consumption of their mothers.

Until recently, complete metabolic model for inhaled mercury vapour in humans was notavailable, despite the fact that the continuous occupational exposure of thousands of workersannually and the long history of mans exposure to this form of mercury. A few studies havepresented toxicokinetics of Hg in experimental animals (Falnoga et al., 1994, Thomas et al.1988). Recently, a four-compartmental model, including two depot compartments to accountfor retention in lungs and kidneys, respectively, gave very good fit of the model output with

OLWUHV[GD\VG[[

& 5 01.0

05.0 95.01-

=

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

17-23

the measurement data (Jonsson et al. 1999). The fraction of dose excreted from the centralcompartment directly into urine was found to be positively correlated with the pre-exposureexcretion rate of Hg via urine. All studies concluded that the true uptake of Hg from amalgamfillings is very difficult to measure due to a number of uncertainties including estimates oneither intraoral measurements, blood or urinary Hg levels. This would of course causedifficulties in modelling exposure of population to Hg in amalgams (Sandborgh, 1998).

7DEOH���� - Mercury concentration in hair in relation to exposure data

&RXQWU\ 7RWDO�+J�LQ�+DLU 5HIHUHQFH

Faroe Islands (Denmark) > 10 µg/g max. 39.1 µg/g (consumptionof pilot whale with 3.3. mg/kg of Hg)

Weihe et al. 1996

Faroe IslandNo fish: 0.8 µg/g1 fish meal/week: 1.6 µg/g2 fish meal/week: 2.5 µg/g3 fish meal/week: 2.1 µg/g4 fish meal/week: 5.2 µg/g

Grandjean et al. 1992

13 countries – a review Once/month: 1.4 µg/g;once/two weeks: 1.9 µg/g;once/week: 2.5 µg/g;once /day: 11.6 µg/g.

Airey et al. 1983

North Sweden 0.28 µg/g Oskarsson et al. 1996Germany 0.25 µg/g Drasch et al. 1997Spain, Madeira Fisherman and their families 38.9 µg/g

(men) and 10.4 µg/g (women)Renzoni 1998

Sweden 1-3 fresh water fish meals/week: 2.5µg/g>3 fresh water fish meals/week: 3.8µg/g1 meal of salt water fish/week: 0.6µg/g

Oskarsson et al. 1990

Bratel et al. 1997

CroatiaUSA 1.5 µg/g Dickman and Leung 1998Brasil

Rio Tapajos

8.76 µg/g (110 g of fish/day with theconcentration of MeHg from 0.1 to 0.5µg/g)26% of fish consumed contain Hg above0.5 mg/kg:High fish diet 16.1 µg/gMixed fish diet: 14.8 µg/gLow fish diet: 7.8 µg/g

Kehrig et al. 1998

Lebel et al. 1997

One may conclude that we still do not have sufficient information to relate mercury levels inair to accumulated body burden and to identify the most appropriate indicator media for levelsof mercury vapour in the target organ (the brain). One study also adressed the farmacokineticsof Hg released from dental amalgams. Inhaled mercury may also cross the placenta, but noinformation is available on human subjects concerning this important question.

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

18-23

���� &21&/86,216�$1'�5(&200(1'$7,216

In order to understand environmental Hg exposures it is necessary to have the information onall sources of Hg exposures.

Quantitative dietary intake data on intakes of all fish and related products should be collectedin any serious study of this contaminant. Such data are necessary to quantify exposures,separating the effect modifiers that account for the differences between exposures and targettissue concentrations. Such data are also essential for identification of possible confoundingfactor such as other contaminant or nutrient that are abundant in this food sources and not inothers.

Mercury in urine is an established marker of long-term exposure to inorganic Hg. Blood andhair analysis provide good estimate of exposure to MeHg through food consumption. In mostsensitive population (e.g. pregnant women and newborn) cord blood Hg concentrations maybe the best surrogate biomarker for Hg in foetal brain compartment.

Exposure assessment in European population, in particular those with high fish consumption,is urgently needed to provide a full picture of the distribution of MeHg and Hg exposuresnationally and regionally. Exposure to elemental Hg from dental amalgams represent aconsiderable daily intake of Hg and should be also considered in risk assessment of MeHg.

Further studies of MeHg exposures in humans should include a thorough assessment of thediet during the period of vulnerability and exposure. Dietary assessments should be conductedconcurrently with the exposure, because retrospective assessment is influenced by manyfactors, including memory, changes in eating behaviour, use of vitamins and mineralsupplementation. In all studies, the estimates should be used with information on MeHgconcentrations in the food to estimate possible MeHg intake by pregnant women, youngchildren and adults. Attempts should be made to validate estimates on the relationshipbetween hair concentration and diet intake. Information on dose from breast-feeding should aswell be improved.

Any biomarker based RfD for MeHg should specifically address interindividual toxicokineticvariability in the estimation of dose corresponding to a given biomarker concentration�

Modelling predictions of individual and population exposure are virtually nonexistent inEurope and should be developed and tested in various environmental conditions and differentsource terms of mercury contamination. To achieve this goal a multidisciplinary approach isneeded linking environmental and health related expertise.

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

19-23

��� 5()(5(1&(6

Airey, D. (1983) Total mercury concentrations in human hair from 13 countries in relation tofish consumption and location. Sci. Total. Env. 31: 157-180.

Barregård L, Sällsten G, Järvholm B. (1995) People with high mercury uptake from their owndental amalgam fillings. Occup Environ Med. ;52:124-128.

Barregård L. (1993) Biological monitoring of exposure to mercury vapor. Scand. J. WorkEnviron. Health 19 suppl 1:45-9.

Barregård L. (1999) Uptake of mercury from dental amalgams - looking beyond the average.In: Novakova V. (ed) Amalgam and Health - New perspectives on risks. Swedish Councilfor Planning Coordination of Research. Stockholm, 1999.

Becker W, Kumpulainen J. Contents of essential and toxic mineralelements in Swedish market-basket diets in 1987. Br. J. Nutr. 66:151-60.

Berglund A.(1990) Estimation by a 24-hour study of the daily dose of intra-oral mercuryvapour inhaled after release from dental amalgam. J. Dent. Res. 69:1646-1651.

Bioschio A.A., Hensel, D.S. (1996) Risk assessment of mercury exposure through fishconsumption by the riverside people in the Madeira Basin, Amazon, 1991. Neurotox., 17:169-175.

Borjesson J, Barregård L, Sällsten G, Schütz A, Jonson R, Alpsten M, Mattsson S. (1995) Invivo XRF analysis of mercury: the relation between concentrations in the kidney and theurine. Phys. Med. Biol. 40:413-426.

Buchet, J.P., Lauwerys, R., Vandevoorde, A., Pycke, J.M. (1983) Oral dietary intake ofcadmium , lead, manganese, chromium, mercury, calcium, zinc, and arsenic in Belgium.A duplicate meal study, Food Chem. Toxicol., 21: 19-24.

Budtz-Jorgensen E, Grandjean P, Keiding N, White RF, Weihe P. (2000) Benchmark dosecalculations of methylmercury-associated neurobehavioural deficits. Toxicol Lett. 112-113:193-9.

Buzina R., Stegnar P., %X]LQD�6XERWLþDQHF�.���+RUYDW�0���Petric I., Farley T.M.M. (1995)Dietray mercury intake and human exposure in an Adriatic population, Sci. Tot. Environ.170: 199-208.

Byrne, A.R. and Kosta, L. (1974) Simultaneous neutron activation determination ofselenium and mercury in biological samples by volatilisation, Talanta 211: 1083.

Clarkson T.W., Friberg, L., Hursh, J.B., Nylander, M. (1988 a) The prediction of intake ofmercury vapour from amalgams. In. Clarkson, T.W., Friberg, L., Nordberg, G.F., Sager,P.R., eds. Biological monitoring of toxic metals, New York, London, Plenum Press,. pp.247-264.

Clarkson, T.W., Hursh, J.B., Sager, P.R., Syversen, T.L.M. (1988b). Mercury, In. Biologicalmonitoring of Biological metals. Eds. T.W. Clarkson, L. Friberg, G.F. Nordberg, P.R.Sager, Plenum Press, New York, pp. 199-246.

Clewell H.J., Gearhard J.M., Gentry P.R., Covington T.R., VanLandingham C.B., CrumpK.S., Shipp A.M. (1999) Evaluation of the uncertainty in an oral reference dose formethylmercury due to inter-individual variability in pharmakokinetics. Risk. Anal. 19(4):547-558.

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

20-23

Craig, P.J. (1986) Organometallic Compounds in the Environment - Principles and Reactions,Longman Group Limited, England.

Dickman, M.D., Leung K.M. (1998) Mercury and organochlorine exposure from fishconsumption in Hong Kong. Chemosphere, 37: 991-1015.

'L]GDUHYLþ�7���������7KH�LQIOXHQFH�RI�PHUFXU\�SURGXFWLRQ�LQ�,GULMD�PLQH�RQ�WKH�HQYLURQPHQWin the Idrija region and over a broad area, Mat. and Geoenviron., 48:, 56-64.

Drasch G., Wanghofer E., Roider G. (1997) Are blood, urine, hair and muscle validbiomonitors for the inetrnal burden of men with heavy metals: mercury, lead andcadmium? Trace elements and Electrolytes, 14: 116-123.

Elinder C.G., Gerhardsson L., Oberdorster (1988) Biological monitoring of toxic metals –Overview. Pp 1-71. In Biological Monitoring of Toxic Metals, T.W. Clarkson, L. Friberg,G.F. Nordberg and P.R. Sager, eds. New York, Plenum press.

EMEP (1999) Monitoring and modelling of lead, cadmium and mercury transboudarytransport in the atmopshere of Europe. Joint report of EMEP Centers: MSC-E and CCC.EMEP report 3/99, July 1999.

EPA Mercury Study Report to the Congress, EPA 452/R-97-0003, EPA, USA, December1997.

Ewers U, Krause C, Schulz C, Wilhelm M. (1999) Reference values and human biologicalmonitoring values for environmental toxins. Int Arch Occup Environ Health 72:255-260.

Falnoga I., Mrhar A., Karba R., Stegnar P., Škreblin M., 7XãHN�äQLGDULþ�0���������0HUFXU\toxicokinetics in Wistar rats exposed to elemental mercury vapour: modeling andcomputer simulations. Arch. Toxicol. 68: 406-415.

Fouassin A and Fondu M. (1978) Evaluation de la teneur moyene en mercure de la rationalimentaire en Belgique. Arch. Belg. Med. Soc. Hyg. Med. Trav. Med. Leg. 36: 481-490.

Gompertz, D. (1982) Biological monitoring of workers exposed to mercury vapour. J. Soc.Occup. Med. 32: 141-145.

Grandjean P., P. Weihe, R.F. White, F. Deves, S. Araki, K. Yokoyama, K. Murata, N.Sorensen, R. Dahl, P.J. Jorgensen (1997), Neurotoxicol. and Teratol�� 1997, ��, 1.

Halbach S.(1994) Amalgam tooth fillings and man’s mercury burden. Human. Exp. Toxicol.13:496-501.

Horvat M. and A.R. Byrne (1992) Preliminary study of the effects of some physicalparameters on the stability of methylmercury in biological samples, Analyst, 117: 665-668.

Horvat M. (1996) Mercury analyses and speciation. In “ Global and Regional MercuryCycles: Sources, Fluxes and Mass Balances” (Edited by W. Baeyens et al.), pp 1-31.Kluwer Academic Publishers, The Netherlands (1996).

Horvat, M, Stegnar, P., Byrne, A.R., Dermelj, M., Branica, Z. (1988) A study of trace lemntsin human placenta, blood, and hair from the Yugoslav Central Adriatic, Trace ElementAnalytical Chemistry in Medicine and Biology, Proc. 5th Int. Workshop, 1988, eds. P.Bratter, P. Schramel, Walter de Gruyter, Berlin, New York, 1988. pp- 243-250.

Horvat, M. and Schroeder, W.H. (1995) Mercury (a) Determination of organomercurials.Encyclopaedia of Analytical Science, Harcourt Brace & Company Limited, London.

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

21-23

Horvat, M. (1999) Current status and future needs for biological and environmental referencematerials certified for methylmercury compounds. Chemosphere (Oxford) 39: 1167-1179.

Huang YL., Cheng SL. and Lin TH. (1996) Lipid peroxidation in rats administrated withmercuric chloride. Biol. Trace Elem. Res. 52:193-206.

Hursh J.B., Cherian M.G., Clarkson, T.W., Vostal J.J., Mallie P.V. (1976) Clearance ofmercury /Hg-197), Hg-203) vapor inhaled by human subjects. Arch. Environ. Health.31(6), 302-309.

IAEA-TECDOC-854, Survey of reference materials, Volume 1: Biological and environmentalreference materials for trace elements, nuclides and micro-contaminants, A reportprepared by the IAEA and UNEP, December 1995, Vienna.

IAEA-TECDOC-880, Survey of reference materials, Volume 2: Biological and environmentalreference materials for trace elements, nuclides and micro-contaminants, A reportprepared by the IAEA and UNEP, May 1996, Vienna.

Ikarashi M., Sasaki K., Toyoda M., Saito Y (1996) Annual daily intakes of Hg, PCB, andarsenic from fish and shellfish and comparative survey of their residue levels in fish bybody weight, Esei Shikenjo Hokoku, 114: 43-47.

IPCS/WHO (1990) Environmental health Criteria 101. Methylmercury. World HealthOrganisation, Geneva, 1990.

IPCS/WHO (1991) Environmental Health Criteria 118. Inorganic mercury, World HealthOrganisation, Geneva, 1991.

IRIS (1995) Mercury, elemental.

IRIS (1995a) Methylmercury.

Ishihara N. and Urushiyama K. (1994) Longitudinal study of workers esposed to mercuryvapour at low concentrations: time course of inorganic and organic mercury concentrationin urine, blood and hair. Occup Environ. Med. 51: 660-662.

Jonsson F., Sandborgh-Englund G., Johansson G. (1999) A compartmental model for thekinetics of mercury vapor in humans. Toxicol. Appl. Pharmacol. 155: 161-168.

Kehrig H.A., Malm O., Akagi H. (1998) Methylamercury in fish and hair samples from theBalbina reservoir, BrazilianAmazon, Environ., Res. 77: 84-90.

Kobal A. and Dizdarevic T. (1997) The health safety programme for workers exposed toelemental mercury at the mercury mines in Idrija. Water Air Soil Pollut. 97 : 169-184.

Kobal, A.B., Osredkar J., Horvat, M., Prezelj, M., Sesek-Briski, A., Kersnik, M., Dizdarevic,T., Falnoga I., Stibilj, (2001) V. The impact of long-term past exposure to elementalmercury on antioxidative capacity and lipid peroxidation in mercury miners, inpreparation.

Kosta, L. and Byrne, A.R. (1969) Activation analyses for mercury in biological samples atnanogram levels Talanta 16: 1297.

Kosta, L., Byrne, A.R., Zelenko, V. 1975. Correlation between selenium and mercury in manfollowing exposure to inorganic mercury, Nature, 254: 238-239.

Kumpulainen J and Tahvonen, R. (1989) Report on the activities of the sub-network on traceelemnts status in food. FAO Report of the 1989 Consultation of the European CooperativeResearch Network on trace elements, Lousanne, september, 1989Amazonian population

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

22-23

LAI, Emissionswerte fur Quecksilber, Quecksilberverbindungen, Berischt desUnterausschusses “Wirkungsfragen”, Berlin: Erich Schmidt Verlag 1996, ISBN 3-503-09364-3.

Lebel.j., Roulet, M., Mergler, D., Lucotte, M., Larribe, F. (1997) Fish diet and mercuryexposure in a riparian areas, Wat. Air Soil Pollut. 97: 31-44.

LECO, AMA 254, Determination of Mercury in Animal Tissues, Organic Application Note,1999. LECO Corporation.

Liang L., and Bloom N. (1993) Determination of total mercury by single amlagamtion coldvapor atomic absorption spectrometry. Analyst, 16, 1297-1301.

Lorscheider FL, Vimy MJ, Summers AO. Mercury exposure from “silver” tooth fillings:emerging evidence questions a traditional dental paradigm. FASEB J 9:504-508.

MAFF, 1994. 1991 Total Diet Study. Food Surveillance Information Sheet 34, July 1994.

Magos, L., (1971) Selective Atomic-absorption determination of inorganic mercury andmethylmercury in undigested biological samples, Analyst 96: 847-852.

Malm O., Calasans C.F., Fernandes A.P., Bastos W.R. and Pfeiffer W.C. Exposure to metallicmercury in workers in a tropical chloralkali plant evaluated through urine analysis, Rio deJaneiro, Brazil. Water Air Soil Poll. 97: 185-191.

Moreiras O., Cuadrado C., Kumpulainen J.T., Carbajal A., Ruiz-Roso B. (1966) Intake ofcontaminants, heavy metals and nutrients with potential toxicity via total diet in fourgeographical areas of Spain. FAO Regional office for Europe, REU Rechnical series 49.Trace elements, natural antioxidants and contaminants in European foods and diets. FAO,Rome, 1996: 59-92.

Nabrzyski, M. and Gajewska, R. (1984) determination of mercury, cadmium, and lead infood. Rocz. PZH, 35(1): 1-11 (in Polish).

Nakagawa R., Yumita Y., Hiromoto M. (1997) Total mercury intake from fish and shellfishba Japanese people, Chemosphere, 35: 2909-2913.

Nordberg G., Skerfving (1972) Metabolism. In. mercury in the environment. A toxicologicaland epidemiological appraisal. L. Friberg and J. Vostal, eds. Chemical Rubber Co.,Cleveland, Ohio. 29-91.

NRC (2000) Toxicological effects of methylmercury Committee on the Toxicological Effectsof methylmercury, Board on Environmental Studies and Toxicology, Commission of LifeSciences, National Research Council, National Academy Press, Washington, DC.

Oskarsson A, Ohlin B, Ohlander EM, Albanus L. (1990) Mercury levels in hairfrom people eating large quantities of Swedish freshwater fish. FoodAddit Contam. 7:555-62.

Bratel J, Haraldson T, Meding B, Yontchev E, Ohman SC, Ottosson JO. (1997)Potential side effects of dental amalgam restorations. (I). An oral andmedical investigation. Eur. J. Oral. Sci. 105:234-43.

Oskarssson A., SchultzA., Skerfving S et al. (1996) Total and inorganic mercury in breastmilk in relation to fish consumption and amalgam in lactating women. Arch. Environ.Health 51: 234-241.

Reimann C. and de Caritat P. (1998) Chemical elements in the environment. Facts heets forthe geochemist and environmental scientist. Springer. 1998.

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

23-23

Renzoni A., ZinoF, Franchi E (1998) Mercury levels along the food chain and risk forexposed populations. Environ. Res. 77: 68-72.

Rodriquez-Vazquez, J.A. (1978) Gas-Chromatographic determination of organomercury(II)Compounds, Talanta 25: 299-310.

Roels H., Abdeladim, S., Ceulemans E. Lauwerys R. (1987) Relationship between theconcentrations of mercury in air and in blood and urine in workers exposed to mercuryvapour. Ann. Occup. Hyg. 31: 135-145.

Ruprich j. 1995. Health risk assessemnt of dietary exposure to the selected chemicalsubstances in the Czech Republic: Alimentary diseases (1993) and total diet study (1994),pp. 274.

Sällsten G, Barregard L, Schütz A. Clearance half life of mercury in urine after the cessationof long term occupational exposure: influence of a chelating agent (DMPS) on excretionof mercury in urine. Occup Environ Med 1994;51:337-342.

Sällsten G, Barregard L, Schütz A. Decrease of mercury concentration in blood after long-term exposure - a kinetic study of chloralkali workers. Br J Ind Med 1993;50:814-821

Sällsten G, Kreku S, Unosson H. (2000) A small dose of ethanol increases the exhalation ofmercury in low exposed humans. J Tox Environ Health. 60:89-100.

Sällsten G, Nolkrantz K. (1998) Determination of trace amounts of mercury vapour in humidair: performance of gold traps in an atomic fluorescence system. Analyst. 123:665-68.

Sällsten G, Thorén J, Barregård L, Schütz A, Skarping G. Long term use of nicotine chewinggum and mercury exposure from dental amalgam fillings. J Dent Res 75:594-598.

Sandborgh Englund G. Pharmacokinetics of mercury from dental amalgam. Thesis. Faculty ofDentistry. Karolinska Institute, Huddinge, Sweden, 1998.

Schroeder W.H. Mercury: Inorganic (and total) determination. In “Encyclopedia of AnalyticalScience”, pp. 3050-3059. Academic Press Limited (1995).

Seigneur C. Pai P., Gerath M., Mitchell D., Hamby G., Gong G., Whipple C., Levin L. (1997)Probabilistic assessment of regional mercury exposure. Wat. Air. Soil Pollut. 97: 159-168.

Skerfving S. (1991) Exposure to mercury in the population. In Advances in MercuryToxicology. Ed. T. Suzuki N. Imura, T.W. Clarkson, , Plenum Press, New York., 1991.411-425.

Sorensen N, Murata K, Budtz-Jorgensen E, Weihe P, Grandjean P. (1999) Prenatalmethylmercury exposure as a cardiovascular risk factor at seven years of age.Epidemiology, 10:370-5.

Stern, A.H., Korn L.R., Ruppel B.E. (1996) Estimation of fish consumption andmethylamercury intake in the new jersey population. J. Expo. Anal. Environ. Epidemiol.6: 503-525.

Steuerwald U, Weihe P, Jorgensen PJ, Bjerve K, Brock J, Heinzow B, Budtz-Jorgensen E,Grandjean P. (2000) Maternal seafood diet, methylmercury exposure, and neonatalneurologic function. J Pediatr. 136:599-605.

Szprengier-Juszkiewicz T. (1988) Evaluation of daily intake of mercury with food stuff inPoland. Bromatol. Chem. Tokykol. 21: 228-232 (in Polish).

Thomas D.J., Fisher H.l., Sumler M.R., Hall L.L., Mushak P (1988) Distribution and retentionof organic and inorganic Hg in methylmercury treated noenatal rats. Env. Res. 47: 59-71.

326,7,21�3$3(5�21�0(5&85<� &+$37(5��

24-23

Tsuda T., Inoue T., Kojima M., Aoki S. (1995) Market basket and duplicate portionestimation of dietary intakes of cadmium, mercury, arsenic, copper, manganese, and zincby japnaese adults, J. AOAC Int. 78: 1363-1368.

Urieta I., Jalon M., EquileroI. (1996) Food surveillance in Basque country (Spain) FoodAddit. Contam. 13: 289-52.

Van Dokkum W., de Vos. R.H., Muys T.H., Westrra, J.A. 1989. Minerals and trace elementsin total diets in the Netherlands, British Journal of Nutrition, 61, 7-15.

Weihe P.D., Grandjean P., Debes F., White R. (1996) Health implications for Feroe islandersof heavy metals and PCBs from pilot whales. Sci. Tot. Env. 186: 141-148.

WHO-FAO (1993) Toxicological evaluation of certain food additives and contaminants.WHO Technical report series No. 837.

WHO (1988) Consultation meeting on the Mediterranean Helth-Related EnvironmentalQuality Criteria (WHO/FAO/UNEP Joint Project, MED POL Phase II). EUR/ICP/CEH059(S), Bled, Slovenia, 12-16. September, 1988.

Woods JS. (1996) Altered porphyrin metabolism as a biomarker of mercury exposure andtoxicity. Can. J. Physiol. Pharmacol. 74(2):210-5.

Yamamura , Y. (1985) Inorganic mercury in blood, urine, and expired air in workers exposedto elemental mercury vapour (in Japanese). Assoc. Ind. Hyg. Jap. 24: 46-51.

Yamamura, Y., Yoshida, M., Yamamura, S., (1991) Blood and urinary mercury levels asindicators of exposure to mercury vapour. In Advances in Mercury Toxicology. Ed. T.Suzuki N. Imura, T.W. Clarkson, Plenum press, New York., 1991. 427-437.

Yoshida, M. (1985) Relation of mercury exposure to elememtary mercury levels in urine andblod. Scand. J., Work. Environ. Health 11: 33-37.